Application of Rejection Sampling based methodology to variance based parametric sensitivity analysis

Liu, Cheng; Lü, Zhenzhou; Zhang, Leigang

doi:10.1016/j.ress.2015.04.020

Cited by 10 publications

(2 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, if

f_{Y, X_{i}} (y, x_{i})

accurately describes the joint PDF of the input‐output sample, we can simply compute various sensitivity indices based on Equation . This has been motivated a number of studies to investigate several strategies for characterizing the joint PDF of the input‐output data for the purpose of variance‐based GSA (see, e.g., Cheng et al., 2015; Jia & Taflanidis, 2016; Sparkman et al., 2016; Wei et al., 2015).…”

Section: Methodsmentioning

confidence: 99%

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

Sheikholeslami

Gharari

Papalexiou

et al. 2021

Water Resources Research

View full text Add to dashboard Cite

Introduction On the High Computational Cost Incurred by the Sampling-Based Global Sensitivity AnalysisWith the rapid increase in computational capability, an increasing number of distributed and semi-distributed process-based hydrologic models have been introduced to simulate the quantity and quality of water on a range of spatiotemporal scales. The ever-growing complexity of these models is driven by the

show abstract

“…Therefore, if

f_{Y, X_{i}} (y, x_{i})

Section: Methodsmentioning

confidence: 99%

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

Sheikholeslami

Gharari

Papalexiou

et al. 2021

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…For simulation-based methods, it is often a challenge to efficiently capture the occurrence of rare events. To address this, several sampling methods, e.g., refined stratified sampling strategy (Shields et al 2015), rejection sampling from predetermined distribution (Cheng et al 2015), and random walks on graphs (Cheng et al 2017) have been proposed. Although the sampling size and efficiency have been improved, challenges still exist in determining the system outcome.…”

Section: Background and Related Workmentioning

confidence: 99%

Probability Propagation Method for Reliability Assessment of Acyclic Directed Networks

Tong

Tien

2019

ASCE-ASME J. Risk Uncertainty Eng. Syst., Part A: Civ. Eng.

View full text Add to dashboard Cite

Many civil infrastructure systems that deliver resources from source points to sinks, e.g., power distribution and gas pipeline networks, can be described as acyclic directed networks comprising nodes and links. Reliability assessment of these systems can be challenging, particularly for systems of increasing size and complexity and if the probabilities of rare events are of interest. This paper proposes a new analytical probability propagation method for reliability assessment of acyclic directed networks called the directed probability propagation method (dPrPm). Through a link-adding sequence to propagate a message consisting of the marginal and pairwise node reliabilities from source nodes to sink nodes, the method results in the upper and lower bounds of all sink node reliabilities. Reliability of a sink node is measured by the probability of reaching that node from a source node. Compared with previous methods, dPrPm addresses the case of multiple-sink networks, results in guaranteed reliability bounds, and analyzes acyclic directed networks as relevant for infrastructure systems. Proofs are provided guaranteeing the accuracy of dPrPm, and computation time is significantly reduced from typical exponential increases with system size to a polynomial increase. To assess performance, the proposed method was applied to three test applications: a directed grid network, a power distribution network, and a more complex gas pipeline network under seismic hazard. Results were compared with the exact solution and Monte Carlo simulations to evaluate accuracy and computational cost. Results showed that dPrPm performs equally well in terms of accuracy across network reliabilities and achieved order-of-magnitude increases in computational efficiency to obtain exact bounds on reliability assessments at all system sink nodes.

show abstract

Interval reliability sensitivity analysis using Monte Carlo simulation and mouth brooding fish algorithm (MBF)

Babazadeh

Rezayfar

Jahani

2023

Applied Soft Computing

View full text Add to dashboard Cite

Application of Rejection Sampling based methodology to variance based parametric sensitivity analysis

Cited by 10 publications

References 54 publications

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

VISCOUS: A Variance‐Based Sensitivity Analysis Using Copulas for Efficient Identification of Dominant Hydrological Processes

Probability Propagation Method for Reliability Assessment of Acyclic Directed Networks

Interval reliability sensitivity analysis using Monte Carlo simulation and mouth brooding fish algorithm (MBF)

Contact Info

Product

Resources

About